[libc-commits] [libc] [libc] Fix missing GPU math implementations (PR #65616)

[Joseph Huber via libc-commits]

jhuber6 wrote:

> The patch is OK but I did not understand the commit message.

Sorry, it's stating that when implemented it seemed like `__builitin_xyz` worked because it did not error, however it was simply being reduced to `xyz` and failing when actually used. This patch pushes them into the "vendor" implementation because they are not actually implemented as builtins. Thus this allows us to use them and also allows the tests to run because we currently ignore the vendor stuff.

https://github.com/llvm/llvm-project/pull/65616